© CAMBRIDGE UNIVERSITY PRESS 1999

1.2 The Discovery

We now return to the observational aspects once again. Even in the early days of radio astronomy, the importance of optical identification of a radio source was appreciated. The identification of Cygnus A had been the key observation to underscore the extragalactic nature of radio sources. Following the optical identification of a suspected extragalactic source, one can hope to do spectroscopy of the optical object and, if possible, determine its redshift. Then, by Hubble's law (see Chapter 2) one can infer the source's distance and hence its radio and optical luminosity.

One of the early catalogues of radio sources was the Third Cambridge Catalogue prepared by the Mullard Radio Astronomy Observatory at the University of Cambridge. Its sources were picked on the basis of their falling within a declination band and being brighter than 9 Jy ¹. The sources were listed in increasing order of right ascension with the prefix 3C. Of these, two sources, 3C 273 and 3C 48, were to play a major role in the developments of the early 1960s.

Radio astronomers at Jodrell Bank were interested in looking at the angular sizes of radio sources. In their first survey they looked at about 300 sources and found that their average size was around 30 arcsec. The size of most sources was in the range of 5 arcsec to a few arcminutes. However, a set of some 10 sources were extremely small, less than 1 arcsec in size. What could they be?

One of them was 3C 48, a source that was optically identified with a star-like object. Allan Sandage found its spectrum to be very unusual, one peculiarity being that it had strong emission lines. Also, Matthews and Sandage (1963) found the light from the object to be variable. These considerations led, in 1962, to this object's being labelled a radio star.

Meanwhile, Cyril Hazard was trying out a new method of fixing the position of a radio source very accurately in the sky. The method, which involved observing the occultation of the source by the Moon, when used on the source 3C 212 seemed to be very promising. And so he wished to try it for the compact source 3C 273, which was also in the Moon's path. Hazard proposed to observe it from Australia along with M.B. Mackey and A.J. Shimmins. The observation was successfully carried out in 1962 and the position of 3C 273 (a source with two components separated by about 20 arcseconds) was obtained with an accuracy of ~ 1 arcsec (Hazard, Mackay and Shimmins 1963).

This positional accuracy was sufficient for astronomers to identify the optical counterpart of the source. It turned out to be a star-like object of some thirteenth magnitude. However, its spectrum, taken by Maarten Schmidt, was peculiar: it had four emission lines. It turned out that they were one of the doublets of [O III], as well as the three hydrogen lines H, H and H. However, all four were appearing with their wave-lengths increased by about 16 per cent. Later, using a spectrum scanner, Beverley Oke also found the expected line H with a similar shift.

Thus the object had a redshift of 16 per cent (see the next chapter for the interpretation of redshift); and as such its original classification as a star in our galaxy went overboard! Instead of being one of the 100 billion or so stars in our galaxy, it had to be an extragalactic object whose redshift-related distance was so large that it was a hundred times brighter than an entire galaxy. Yet, it had to be compact enough to be mistaken for a star. Further, examination of old plates showed that the light from this object had also fluctuated significantly with time. Figure 1.1 illustrates the optical object with the optical jet clearly seen at its lower right-hand side.

The radio source structure of 3C 273 determined by lunar occultation showed it to be a two-component system with the components separated by 19.5 arcsec. Of the two components A and B, the optically identified object sits on component B and shows a jet directed towards the other component, A. The jet is another indication that the object is not a star but is probably a much more violent system. The radio structure on various angular scales can be seen in Figure 9.13.

Fig. 1.1. The quasar 3C 273 in the optical band. Reproduced from Narlikar (N93).

Alerted by these unusual features in 3C 273, astronomers took a second look at the object 3C 48 and found that its spectrum had looked unusual (or, unfamiliar) because it too was redshifted, and by an even larger degree, around 36.7 per cent.

It should be recalled that in the early 1960s the measured redshifts of galaxies usually ranged up to about 0.2, i.e., 20 per cent. The galaxy identified with the radio source 3C 295 had a redshift of 0.46, which was then the record! Thus the large redshifts of these two unusually compact objects immediately drew the attention of theoreticians. Further, this discovery, coming as it did in early 1963, coincided with the theoretical expectations of Hoyle and Fowler that highly compact massive objects could serve as high energy reservoirs.

The role of theoretical inputs was considered so important that an international symposium was convened in Dallas, Texas in December 1963 to bring together general relativists, theoretical astrophysicists and observers to discuss the implications of the discovery of these remarkable objects. This was to be the first of a continuing series of biennial symposia under the name ``Texas Symposia''. The objects became known as quasi-stellar objects (QSO) or quasars. we will use either of these names in this text. ⁽²⁾

² See the discussion in Section 6.1.